Thesis Defense Seminar - Rostam E. Mirzadi
6300 Ocean Drive, Corpus Christi, TX 78412
DEPARTMENT OF LIFE SCIENCES
TEXAS A&M UNIVERSITY-CORPUS CHRISTI
SUBJECT: Evaluating the Use of Drones for Monitoring Waterbird Nest Abundance and Nest Survival
MAJOR ADVISOR: Dr. Dale Gawlik
Surveys of colonial waterbirds are used to monitor species’ population status, contaminant levels, and to derive metrics that can be used to assess wetland ecosystem restoration and management and activities. Previous studies have found that drone surveys provide accurate estimates of nest abundance and survival for ground-nesting waterbird species such as terns (Laridae Spp.), but drones have not been used to estimate survival for waterbirds nesting in a canopied marsh habitat, and potential sources of bias in drone surveys have not been examined in depth. I examined potential visibility biases associated with using a drone to survey colonies of wading birds (Ciconiiformes and Pelecaniformes) in marsh habitat in Florida in 2020 and 2021. Monthly nest counts and survival were compared between traditional (combination of fixed-wing aircraft and ground surveys) and drone survey methods. Ground-based and drone nest transect surveys were conducted to estimate survival and detection probabilities of each species and plumage color. Generalized linear mixed-effects models were used to quantify the degree to which visual occlusion of nests inﬂuenced detectability. Estimates of white-colored waterbird nests were significantly greater for drone surveys than those derived from traditional survey methods but estimates of dark-colored waterbirds from drone surveys were biased low. Variation in detection was best explained by canopy cover, plumage color, and nest stage. Overall, there was no difference between survival estimates from either method. However, drone-derived estimates of dark-colored waterbirds had lower accuracy. Our results suggest that drone surveys are a viable method to conduct monthly nest surveys and estimate survival of waterbirds breeding in marsh habitat, but researchers should consider their study area and species because before choosing a survey method.
Meeting ID: 351 826 4930